HEPARIN POTENTIATES THE IN VIVO ECTOPIC BONE FORMATION INDUCED BY BONE MORPHOGENETIC PROTEIN-2

Although bone morphogenetic proteins (BMPs) are clinically useful for bone regeneration, large amounts are required to induce new bone formation in monkeys and humans. We found recently that heparin stimulates BMP activity in vitro (Takada, T., Katagiri, T., Ifuku, M., Morimura, N., Kobayashi, M., Hasegawa, K., Ogamo, A., and Kamijo, R. (2003) J. Biol. Chem. 278, 43229-43235). In the present study, we examined whether heparin enhances bone formation induced by BMPs in vivo and attempted to determine the molecular mechanism by which heparin stimulates BMP activity using C2C12 myoblasts. Heparin enhanced BMP-2-induced gene expression and Smad1/5/8 phosphorylation at 24 h and thereafter, although not within 12 h. Heparitinase treatment did not affect the response of cells to BMP-2. In the presence of heparin, degradation of BMP-2 was blocked, and the half-life of BMP-2 in the culture medium was prolonged by nearly 20-fold. Although noggin mRNA was induced by BMP-2 within 1 h regardless of the presence of heparin, noggin failed to inhibit BMP-2 activity in the presence of heparin. Furthermore, simultaneous administration of BMP-2 and heparin in vivo dose-dependently induced larger amounts of mineralized bone tissue compared with BMP-2 alone. These findings clearly indicate that heparin enhances BMP-induced osteoblast differentiation not only in vitro but also in vivo. This study indicates that heparin enhances BMP-induced osteoblast differentiation in vitro and in vivo by protecting BMPs from degradation and inhibition by BMP antagonists.


Introduction
Bone morphogenetic proteins were originally identified as an activity inducing ectopic bone formation when implanted into muscle tissue (1).
With the exception of BMP-1, BMPs are members of the transforming growth factor-(TGF-) superfamily, which includes multifunctional factors in vertebrates and invertebrates (3).
BMPs regulate the differentiation and function of cells that are involved in bone and cartilage formation and deformation, including osteoblasts, chondrocytes, and osteoclasts (4,5,7). A cell culture system using C2C12 myoblasts can mimic part of the process of the ectopic bone formation induced by BMPs in muscle tissue (8).
BMPs inhibit myogenic differentiation of myoblasts into mature muscle cells and convert their differentiation pathway into that of osteoblast lineage cells. Using this model system in vitro, a critical factor for osteoblast differentiation, termed osterix, has been cloned as a novel Zn finger transcription factor induced by BMP-2 (9). Osterix knockout mice do not have bone tissues due to lack of osteoblast differentiation from mesenchymal progenitor cells (9). Although several growth factors including TGF-and activin inhibit the myogenic differentiation of C2C12 cells in vitro, they do not induce osteoblast differentiation (8,10). It has been reported that TGF-itself failed to induce ectopic bone formation in vivo (2). These findings suggest that the C2C12 cell model system is useful for examining the molecular mechanisms of the TGF-superfamily members during ectopic bone formation in muscle tissue. It should therefore be useful for examining the molecular mechanism of bone formation and osteoblast differentiation induced by BMPs.
BMP signaling is transduced by two types of transmembrane serine/threonine kinase receptors.

Recombinant proteins and heparin
Purified recombinant human BMP-2 was obtained from Astellas Pharmaceuticals Co., Ltd. (Tokyo, Japan cDNA was denatured at 94º C for 5 min followed by repeated 25 cycles of 45 sec at 94º C, annealing at 55 º C for 1 min, and extension at 68 º C for 1 min.

Histochemical analysis
To determine protein levels immunohistochemically, cells were incubated with appropriate primary and secondary antibodies as described (8). were stained for enzyme activity as described (8).
Staining was visualized with a BX50 fluorescence microscope with a DP50 digital CCD camera (Olympus, Tokyo, Japan). according to the manufacturer's instructions.

Ectopic bone formation assay
The effects of heparin on bone formation in vivo induced by BMP-2 were examined by ectopic bone formation assay in mice (24). week, and they were divided into 4 groups (8 mice per group). The mice were anesthetized by diethyl-ether gas inhalation, and the collagen pellets prepared as described above were surgically implanted into the both dorsal muscle pouches (two pellets per animal) of the mice. At 2, 3, and 4 weeks after surgery, the mice were sacrificed, and the implants were harvested and processed for histological analysis as follows.
All harvested samples were radiographed with a soft X-ray apparatus (Sofron Co., Ltd., Tokyo, Japan

Specificity of enhancing effect of heparin on induction of ALP activity by BMPs
We first compared the effect of heparin on ALP activity, a typical marker of osteoblast differentiation, induced by BMP-2, BMP-4, and BMP-6 in C2C12 cells.
As we reported previously, heparin at 5 mg/ml enhanced the ALP activity induced by both BMP-2 and BMP-4 (Figs. 1A and 1B). In contrast, the enzyme activity induced by BMP-6 was suppressed by heparin ( Fig. 1C). Drosophila DPP, a homolog of mammalian BMP-2 and BMP-4, failed to induce ALP activity in the presence or absence of heparin in our assay system (data not shown), though highly purified heparin had effects on BMP-2 similar to those of the heparin used throughout the present study (Fig. 1D). Low-molecular-weight (LMW) heparin, which is roughly 5-fold smaller than native heparin, also enhanced induction of ALP activity by BMP-2, but to a lesser extent than native heparin (Fig. 1D). The estimated ED 50 of LMW heparin and native heparin were 10 and 2 mg/ml, respectively, suggesting that the size of We previously reported that heparin did not enhance the osteoblast differentiation induced by activation of a signaling pathway downstream from BMP receptors, suggesting that a primary target of heparin is upstream of the receptor. In agreement with this finding, pretreatment of C2C12 cells with heparin from day -1 to day 0 did not enhance the ALP activity induced by BMP-2 on day 3 (Fig. 2C). Moreover, pre-incubation of BMP-2 with heparin before treatment did not affect ALP activity (Fig. 2C). Although we can not rule out the possibility that the pre-incubation with heparin was negligible on day 3, these results suggest that heparin rapidly modified BMP-2 in culture media.

BMP-2 activity
We further examined the time-courses of change in levels of expression of genes related to osteoblast differentiation, such as ALP, osteocalcin, osterix, and Runx2 type I and types II/III (Fig. 3). Levels of Runx2 type I were not changed by BMP stimulation either with or without heparin until 48 h. The expression of the other genes was induced by BMP-2 and further stimulated by heparin after 24 h. However, the extent of induction of osterix by BMP-2 at 4 h was not enhanced by heparin (Fig. 3).
We was also increased by heparin in a dose-dependent fashion (Fig. 7C). At 4 weeks, the inside of the bone pellets treated with BMP-2 alone was filled with bone marrow. In the presence of heparin, however, not only bone marrow but also much trabecular bone was observed in the pellets (Fig.   7D).  for the BMP-2 and BMP-7 subgroups (4, 5).

Discussion
These issues require further examination.
Heparin is thought to increase the binding affinity of FGF to its specific receptor and thereby enhance downstream signaling (18,19).